kernel&chai
Start a project

services

We build the hard parts and ship the whole thing.

Senior engineering for startups and teams: product web apps, products that AI assistants can use safely, 3D in the browser, and the cloud it all runs on.

01 / 04

Web apps & SaaS

Full products from first prototype to paying customers: front end, back end, auth, billing, dashboards.

  • React + TypeScript front ends
  • .NET or Node back ends, REST and real-time (SignalR, WebSockets)
  • Auth, roles, multi-tenant data, billing
  • Admin panels and internal tools

02 / 04

Agent-ready products

Let AI assistants use your product safely, and ship AI features that hold up in production.

  • MCP servers so Claude and other assistants can use your product
  • Scoped tools: each user’s assistant sees only what their role allows
  • Writes off by default, human approval where it matters, audited calls
  • AI features: agents, retrieval, tool use, cost control

03 / 04

3D on the web

Heavy 3D in a browser tab: scans, point clouds, Gaussian splats and models that load fast on any device.

  • Three.js / WebGL viewers and editors
  • Point clouds and LiDAR scans, streamed
  • Gaussian splats and photogrammetry
  • Product, space and site visualisation

04 / 04

Cloud & delivery

Infrastructure that ships every week and does not wake you up at night.

  • AWS and Cloudflare
  • CI/CD, staging, previews, rollbacks
  • Performance, scaling and cost reviews
  • Monitoring and on-call handover

start small

Fixed-price starter offers.

A short engagement with a clear deliverable. If the answer is "don't build this", you hear it here, before you've spent the big budget.

1 week fixed price

Agent-readiness review

We review how AI assistants could use your product, or audit the MCP server you already have: tool design, permissions, write safety, data exposure.

you getRanked findings and a build plan

Ask about this →
2 weeks fixed price

3D viewer sprint

Your scans, models or splats running in a browser on your data, with real load times on real devices. Ends in a go / no-go you can act on.

you getWorking prototype and a feasibility report

Ask about this →
5 business days fixed price

AI app production audit

A fixed-price review of an AI feature you already shipped: cost per conversation, caching, security, prompt-injection exposure, failure handling.

you getRanked report with fixes

Ask about this →

engagement

Ways to work together.

Fixed-scope project

A defined product or feature, a fixed quote, weekly releases you can click through.

Monthly retainer

A senior engineer on your product every month: roadmap, builds, reviews, upkeep.

Starter offer

One of the fixed-price engagements above. A low-risk way to see how we work before a bigger build.

process

How a project runs.

  1. 01

    Discovery call

    30 minutes, free. What you are building, for whom, and what done looks like.

  2. 02

    Scope & quote

    A written plan: features, architecture, timeline and a fixed price or monthly rate.

  3. 03

    Build in the open

    Weekly releases to a staging link. You see progress, not status reports.

  4. 04

    Launch & hand over

    Production deploy, docs, and the code in your repository. Support if you want it.

faq

Questions, answered.

Anything else? Ask directly.

Do you work with early-stage startups?

Yes. A lot of the work is taking a product from idea or rough prototype to something customers pay for. We will tell you what to cut to launch sooner.

Who owns the code?

You do. Everything lives in your repository and accounts from day one.

How do you price?

Fixed quotes for defined projects, a monthly rate for retainers, and a fixed price for audits. You get the number in writing before any work starts.

Where are you based?

India, working with teams anywhere. Calls overlap with European and US mornings.

next step

Have something to build?

Tell us what you are making. You get a straight answer on scope, approach and cost after a free 30-minute call.

Start a project → Book a call
esc